Archiving source transcription files

Identifying source transcription files

If you are uploading a transcript of your voice file from the source platform, e.g. Zoom, you can specify whether the file is a transcription. Once archived, you can identify messages with transcription files using the X-GR-Transcription x-header.

If you enabled Global Relay transcription for the data type you are archiving via the Voice Archiving API, any supported voice files you include in your requests will be automatically transcribed as separate files in your archived messages. You can search for Global Relay transcription files using the general X-GR-Transcription header as well as the Global Relay transcription-specific X-GR-TranscriptionVersion header.

After uploading the relevant source transcription file via the /files endpoint, you can specify the voice file in the voiceRecords property and the transcription file in the extraFiles property.

You can then indicate the file was a transcript in the following fields of the extraFiles property:

  • isTranscript: Indicate whether the file is a transcript file, i.e. true. By default, the value is set to false.
  • description: Indicate the file was generated via the source platform, e.g. “Transcript file generated via source platform”
"voiceRecords": [
    {
      "startTime": "1737050060753",
      "stopTime": "1737050150341",
      "recordingFile": {
        "filename": "meeting-recording-20250115.opus",
        "fileKey": "/default/meeting-recording-20250115.opus",
        "description": "Recording of conversation"
       },
    }
  ],
"extraFiles": [
        {
            "filename": "meeting-recording-20250115.ttl",
            "fileKey": "default/meeting-recording-20250115.ttl",
            "description": "Transcript file generated by source platform",
            "isTranscript": true
        }
    ]
}